From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753196AbcGLGos (ORCPT ); Tue, 12 Jul 2016 02:44:48 -0400 Received: from mx1.redhat.com ([209.132.183.28]:43411 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750743AbcGLGoq (ORCPT ); Tue, 12 Jul 2016 02:44:46 -0400 Date: Tue, 12 Jul 2016 08:49:43 +0200 From: Alexander Gordeev To: Christoph Hellwig Cc: tglx@linutronix.de, axboe@fb.com, linux-block@vger.kernel.org, linux-pci@vger.kernel.org, linux-nvme@lists.infradead.org, linux-kernel@vger.kernel.org Subject: Re: [PATCH 08/13] pci: spread interrupt vectors in pci_alloc_irq_vectors Message-ID: <20160712064943.GB26968@agordeev.lab.eng.brq.redhat.com> References: <1467621574-8277-1-git-send-email-hch@lst.de> <1467621574-8277-9-git-send-email-hch@lst.de> <20160707110501.GA4917@dhcp-27-118.brq.redhat.com> <20160710035751.GC15720@lst.de> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20160710035751.GC15720@lst.de> User-Agent: Mutt/1.5.21 (2010-09-15) X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.5.16 (mx1.redhat.com [10.5.110.26]); Tue, 12 Jul 2016 06:44:46 +0000 (UTC) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Sun, Jul 10, 2016 at 05:57:51AM +0200, Christoph Hellwig wrote: > On Thu, Jul 07, 2016 at 01:05:01PM +0200, Alexander Gordeev wrote: > > irq_create_affinity_mask() bails out with no affinity in case of single > > vector, but alloc_descs() (see below (*)) assigns the whole affinity > > mask. It should be consistent instead. > > I don't understand the comment. If we only have one vector (of any > kinds) there is no need to create an affinity mask, we'll leave the > interrupt to the existing irq balancing code. I got impression the affinity mask is assigned differently i.e. for single MSI mode vs legacy mode. But I might be wrong here - I have to look into the code again. Anyway, if it is the case it could be addressed afterwards IMHO.